Jari Arkko's intervention in the "Dynamic Coalition for Internet of
Things" session at IGF 2016, Guadalajara, Mexico, December 2016.

First, I had a couple of comments on the statements and questions.

I do of course agree with the importance of the balance in finding the
ethical approach. And I agree with the need to educate users, and
frankly, the industry and developers in some areas as well.

And definitely agree on the need for ontologies. Also agree on proper
security to avoid capture of devices. More on those topics in a sec.

With regards to the stimulating the use of the most available
technology possible, I'm not so sure on that. Technology adoption is
driven by need. But let me take an another angle on this.  In this
whole conference, we have once again talked at length on connecting
the people who are not connected. That is of course very
important. And much work remains. But it is NOT enough.

There's much more that we need to do. Not just a matter of high-speed
broadband. A question of quality, not just quantity. The open
Internet, local content, the ability of everybody to create new
services. And on the IOT front, what about say instrumenting farming
fields or acquiring the devices that can help people's health?
Clearly, we have lots to do in the developing parts of the world, even
after everyone is connected.

But I wanted to get back to the deeper technical aspects. I want to
highlight two problems that we at the IETF and Internet Architecture
Board have worked with this year. First, I'm very happy about how
we're building interoperable systems and standards, different devices
can exist in the same generally useful networks, we can communicate
over IP to other devices around the world, and so on.

But unfortunately, we have ways to go when it comes to
interoperability at the application level. I don't want to buy
a house with microsoft light switches and then find out
I can't plug in apple lightbulbs. More work needed, and I
wish the coalition paper said more of this. This is where
ontologies, semantic descriptions, and more standards
work comes in.

Secondly, as you've seen recently, the problems relating to
the security of the IOT devices are mounting. The situation
is bad, and it is not just about devices and their users being
affected, it is also devices being hijacked to cause havoc
somewhere else.

Work is obviously needed here, and it will need to go into
diverse topics, both technical, policy, legal, liability, and so
on. This is a perfect thing for the IGF in general to do, and
also wish to see some additions in the coalition paper. From
a techie perspective I wonder if we need to write a baseline
requirements RFC that says you can't deploy millions of
devices with default passwords. Duh.

Thank you.